Digital learning technologies create new opportunities for personalising the learning process. By personalising each learner’s environment, we can better align it to their needs. When thinking about personalisation in this context, there are several important questions to consider. Which elements of a learning environment can be personalised? And who decides what to personalise and how?
Who decides?
Digital learning environments create an interaction between a human user and a digital machine. Since human and machine work together, the locus of control over aspects of such a digital environment can vary: the learner can be in full control, share control with the digital system, or the system can have full control. For example, the scheduling of practice items within a learning session may be controlled by a learner, but in an adaptive learning system like MemoryLab is usually controlled by the computer, based on the premise that this decision is better taken out of the learner’s hands. The same spectrum of control exists when it comes to personalisation of the learning environment. On one end of the spectrum, a learner may have full say over how the environment is configured; on the other end, personalisation may happen fully automatically. In the former case, the system can be said to be adaptable, while in the latter case, the system is adaptive (Plass and Pawar, 2020). How do we strike the right balance?
Often, a mix of adaptability and adaptivity is the right answer. Automatic personalisation of the learning environment can be convenient, as it reduces the number of choices the learner is asked to make. Automatic personalisation can also help steer a learner towards more effective learning practices that they otherwise may not seek out. For instance, when it is clear that a learner has advanced sufficiently to practise at a higher difficulty level, the learning environment can automatically make this change on behalf of the learner. When left to their own devices, learners can make poor decisions about how they configure their learning environment, for example as a result of limited or flawed insight into their own learning process and abilities (Metcalfe, 2009, Bjork, Dunlosky, & Kornell, 2013). In such cases, automatic personalisation may be advantageous. If the digital learning environment already uses a learner model to automatically adapt the content to the learner’s performance, automatic personalisation can also be integrated into that existing model (Peng, 2019).
At the same time, there are clear benefits to giving learners the opportunity to personalise a digital environment themselves (i.e., making the environment more adaptable). Developing learners’ agency is often an important educational goal in itself (Stenalt, 2021). When a learner has a greater sense of control, they are more likely to experience interest in the learning task, be more motivated to perform well, and are likely to enjoy the task more (Ryan and Deci, 2000). For example, a recent study by Wang et al. (2015) found that second-language vocabulary learners were more motivated and engaged when they could choose their own target words to study, rather than studying a prescribed set of items. Alongside such motivational and affective benefits, promoting learners’ agency has also been found to improve learning outcomes like skill mastery or retention of knowledge. For instance, Long and Aleven (2016) demonstrated that giving mathematics learners a say in the selection of practice problems led to better mastery of the material than having the selection of problems be fully machine-driven.
What can be personalised?
Personalisation in an adaptive learning environment like MemoryLab can be about many different elements of the learning experience. In the case of MemoryLab, there are many opportunities for personalisation. Here is a (non-exhaustive) list of some of these.
- The configuration of the interface: What do components of the learning interface look like to the learner? Which components can be added, and which can be hidden from view?
- The manner in which the learner interacts with the system: What is the format of the question? How does the learner provide an answer? How quickly or precisely does the learner need to respond?
- The structuring of learning material: Which items does a learner study? How many items are there in a session? How and when are new items introduced?
- The configuration of the adaptive learning algorithm: At which difficulty level does the learner practice? What is the performance level that the learner needs to achieve? How does the system schedule items within and between learning sessions?
- The role of automated analytics in the learning process: How do model-based analytics like Model-Based Mastery assessment feature in the learning process? Which automatic study recommendations does the system make?
At the moment, personalisation of many of these features does not yet exist in MemoryLab. As part of a collaboration with ETS EMEA and the University of Groningen and with funding from Samenwerkingsverband Noord-Nederland, we are currently researching how to implement more personalisation within the system. In this project, our focus lies on personalisation of features that benefit neurodivergent and low-performing learners, though other learners may also benefit. Our aim is to achieve a learning environment that better suits the needs of all learners, helping them both to demonstrate what they know and to acquire new knowledge more effectively.
References
Bjork, R. A., Dunlosky, J., & Kornell, N. (2013). Self-Regulated Learning: Beliefs, Techniques, and Illusions. Annual Review of Psychology, 64(1), 417–444. https://doi.org/10.1146/annurev-psych-113011-143823
Long, Y., & Aleven, V. (2016). Enhancing learning outcomes through self-regulated learning support with an Open Learner Model. User Modeling and User-Adapted Interaction, 27(1), 55–88. https://doi.org/10.1007/s11257-016-9186-6
Metcalfe, J. (2009). Metacognitive Judgments and Control of Study. Current Directions in Psychological Science, 18(3), 159–163. https://doi.org/10.1111/j.1467-8721.2009.01628.x
Peng, H., Ma, S., & Spector, J. M. (2019). Personalized adaptive learning: An emerging pedagogical approach enabled by a smart learning environment. Smart Learning Environments, 6(1), 9. https://doi.org/10.1186/s40561-019-0089-y
Plass, J. L., & Pawar, S. (2020). Toward a taxonomy of adaptivity for learning. Journal of Research on Technology in Education, 52(3), 275–300. https://doi.org/10.1080/15391523.2020.1719943
Ryan, R. M., & Deci, E. L. (2000). Self-Determination Theory and the Facilitation of Intrinsic Motivation, Social Development, and Well-Being. American Psychologist.
Stenalt, M. H. (2021). Digital student agency: Approaching agency in digital contexts from a critical perspective. Frontline Learning Research, 9(3), 52–68. https://doi.org/10.14786/flr.v9i3.697
Wang, H.-C., Huang, H.-T., & Hsu, C.-C. (2015). THE IMPACT OF CHOICE ON EFL STUDENTS’ MOTIVATION AND ENGAGEMENT WITH L2 VOCABULARY LEARNING. Taiwan Journal of TESOL, 12(2), 1–40.
Woodfine, B. P., Nunes, M. B., & Wright, D. J. (2008). Text-based synchronous e-learning and dyslexia: Not necessarily the perfect match! Computers & Education, 50(3), 703–717. https://doi.org/10.1016/j.compedu.2006.08.010
